Anthropic CCDV-F (Claude Certified Developer - Foundations) Practice Test 31

Question 300

The current design of a Claude Code repository pipeline does not adequately apply specialised rules only to a sensitive infrastructure folder. Which improvement resolves the gap?

  1. Use a project-level CLAUDE.md file
  2. Use headless invocation for automation
  3. Connect an MCP server to Claude Code
  4. Use directory-scoped instructions

Explanation

Correct Answer

D:

More local instruction files help scope guidance to the appropriate subtree; this behavior directly supports the need to apply specialised rules only to a sensitive infrastructure folder. Root guidance alone can be too broad when different modules have conflicting conventions.

Incorrect Answers

A:

A checked-in memory file supplies persistent project context across relevant coding sessions; the supported outcome is to share coding conventions with every collaborator entering the same repository, whereas this scenario requires the ability to apply specialised rules only to a sensitive infrastructure folder; a personal user-level instruction file does not automatically impose a shared team standard.

B:

This approach serves the separate need to run Claude Code inside a CI job and capture its output without interactive prompts because noninteractive execution lets scripts integrate agent work into existing pipelines with controlled formats; it does not establish the outcome to apply specialised rules only to a sensitive infrastructure folder, and treating interactive terminal output as a stable machine interface is fragile.

C:

An MCP integration declares external capabilities and their transport while access remains controlled by the server; despite being useful when a service must expose approved project-specific actions without copying credentials into prompts, it is not a solution to apply specialised rules only to a sensitive infrastructure folder; a server connection does not automatically make every operation safe or authorized.

Question 301

A release plan for an inference routing platform has two acceptance checks: component one must clarify a specific formatting convention that a plain instruction was not conveying; component two must shorten latency and spend on tasks that need only basic categorization. Which proposal passes both?

  1. First component: Use varied sampling only for ideation. Second component: Reduce reasoning effort on trivial routing.
  2. First component: Use one representative example. Second component: Reduce reasoning effort on trivial routing.
  3. First component: Use one representative example. Second component: Use low-temperature generation for repeatable extraction.
  4. First component: Use low-temperature generation for repeatable extraction. Second component: Use varied sampling only for ideation.

Explanation

Correct Answer

B:

The first proposal fits its assigned role because a single demonstration can anchor the expected shape and tone of a generated response, while the second proposal fits its assigned role because lower effort can suit predictable work where deeper reasoning adds little measurable quality; their respective acceptance criteria are to clarify a specific formatting convention that a plain instruction was not conveying and to shorten latency and spend on tasks that need only basic categorization.

Incorrect Answers

A:

The first proposal serves the different purpose to brainstorm diverse naming candidates without demanding identical wording on every run instead of the assigned first role because greater sampling diversity can produce a wider range of suggestions for exploratory work, while the second proposal fits its assigned role because lower effort can suit predictable work where deeper reasoning adds little measurable quality; their respective acceptance criteria are to clarify a specific formatting convention that a plain instruction was not conveying and to shorten latency and spend on tasks that need only basic categorization.

C:

The first proposal fits its assigned role because a single demonstration can anchor the expected shape and tone of a generated response, while the second proposal serves the different purpose to reduce variation when a service maps documents into consistent categories instead of the assigned second role because lower randomness often improves consistency for narrow, constrained tasks; their respective acceptance criteria are to clarify a specific formatting convention that a plain instruction was not conveying and to shorten latency and spend on tasks that need only basic categorization.

D:

The first proposal serves the different purpose to reduce variation when a service maps documents into consistent categories instead of the assigned first role because lower randomness often improves consistency for narrow, constrained tasks, while the second proposal serves the different purpose to brainstorm diverse naming candidates without demanding identical wording on every run instead of the assigned second role because greater sampling diversity can produce a wider range of suggestions for exploratory work; their respective acceptance criteria are to clarify a specific formatting convention that a plain instruction was not conveying and to shorten latency and spend on tasks that need only basic categorization.

Question 302

A technical lead requests a solution that can keep integration behavior aligned with a supported request contract within a model platform project. Which design should be chosen?

  1. Specify request deadlines in the caller
  2. Send an API version header where required
  3. Use the Messages count_tokens capability
  4. Honor server rate-limit signals

Explanation

Correct Answer

B:

Explicit protocol versioning makes the caller's expected API semantics unambiguous; this behavior directly supports the need to keep integration behavior aligned with a supported request contract. Relying on an arbitrary undocumented default is fragile.

Incorrect Answers

A:

This approach serves the separate need to keep a web request from hanging indefinitely if a network path stalls because a client timeout bounds waiting for transport and response completion; it does not establish the outcome to keep integration behavior aligned with a supported request contract, and model generation settings alone cannot enforce a socket deadline.

C:

For the alternate objective to estimate model-token input demand before launching a large processing job, token counting helps check input budgets before generation and cost planning; the feature here must keep integration behavior aligned with a supported request contract, and word or byte count is not a reliable substitute across diverse content.

D:

Throttling and backoff work with provider limits rather than assuming infinite request capacity; this applies to the different goal to prevent a background processor from overwhelming the upstream API during bursts, not the requirement to keep integration behavior aligned with a supported request contract; parallelism beyond quota can reduce successful throughput.

Question 303

During acceptance testing for a document processing service, one requirement is to distinguish an invalid request from a temporary provider or rate-limit failure. What should the developer implement?

  1. Use supported document or PDF input
  2. Include an image content block
  3. Handle the max_tokens stop_reason
  4. Read the HTTP error status before retrying

Explanation

Correct Answer

D:

Status codes and structured errors guide correction versus bounded retry; this behavior directly supports the need to distinguish an invalid request from a temporary provider or rate-limit failure. Blind retries of permanent validation errors waste time and quota.

Incorrect Answers

A:

This approach serves the separate need to extract claims from a document while retaining its visual layout because document-aware input lets a capable model consider page content and graphics when supported; it does not establish the outcome to distinguish an invalid request from a temporary provider or rate-limit failure, and pasting only extracted text can discard useful visual context.

B:

Image inputs are supplied as typed content with appropriate source data or references; despite being useful when a service must analyze a photographed shipping label with the supported multimodal Messages input, it is not a solution to distinguish an invalid request from a temporary provider or rate-limit failure; describing a local image pathname in ordinary text does not transmit its pixels.

C:

For the alternate objective to recover a response that ended because its output budget was exhausted, the finish reason identifies truncation, so a caller may need a continuation or a revised token budget; the feature here must distinguish an invalid request from a temporary provider or rate-limit failure, and treating that response as a complete answer can hide missing fields.

Question 304

A technical lead requests a solution that can preserve customer preferences across permitted sessions while distinguishing facts from assumptions within an engineering productivity project. Which design should be chosen?

  1. Use a graceful fallback path
  2. Use bounded memory with explicit provenance
  3. Adopt a blackboard of shared task state
  4. Delegate narrow subtasks with a separate context

Explanation

Correct Answer

B:

Typed, attributable records prevent outdated or unverified memories from becoming unquestioned instructions; this behavior directly supports the need to preserve customer preferences across permitted sessions while distinguishing facts from assumptions. Arbitrary model-generated memories can become a source of drift.

Incorrect Answers

A:

A bounded failure branch can return a safe partial result or route to a human rather than loop indefinitely; this applies to the different goal to keep a customer service application operating when an expensive specialist agent cannot complete, not the requirement to preserve customer preferences across permitted sessions while distinguishing facts from assumptions; unbounded retries do not ensure eventual success.

C:

A controlled shared workspace holds versioned observations and ownership metadata for collaborators; the supported outcome is to coordinate multiple specialists working on interdependent evidence without losing changes, whereas this scenario requires the ability to preserve customer preferences across permitted sessions while distinguishing facts from assumptions; unstructured message passing alone can create stale assumptions.

D:

This approach serves the separate need to protect a coordinator's window from huge specialist tool outputs because each subagent can work in a bounded context and return a concise, verified artifact; it does not establish the outcome to preserve customer preferences across permitted sessions while distinguishing facts from assumptions, and copying every intermediate trace into the parent inflates context and cost.

Question 305

A team maintaining an output validation needs to avoid answering a pricing question from last month's retrieved cache. Which implementation most directly satisfies the requirement?

  1. Refresh stale evidence before reuse
  2. Deduplicate overlapping passages
  3. Preserve instruction and evidence boundaries
  4. Anchor summaries to source identifiers

Explanation

Correct Answer

A:

Freshness checks determine whether a saved passage remains authoritative for a dynamic fact; this behavior directly supports the need to avoid answering a pricing question from last month's retrieved cache. High semantic similarity says nothing about timeliness.

Incorrect Answers

B:

Overlap detection reduces redundant context without throwing away unique clauses; this applies to the different goal to save tokens when several retrieval chunks contain the same repeated policy section, not the requirement to avoid answering a pricing question from last month's retrieved cache; simply increasing retrieval count may provide many copies of the same material.

C:

When an application needs to keep an embedded webpage instruction from becoming higher-priority application policy, context assembly must label external text as data and avoid granting it authority over tool use; the current capability must instead avoid answering a pricing question from last month's retrieved cache, while quotation formatting alone does not enforce sandboxing or authorization.

D:

Stable document IDs and passage references preserve provenance beyond the shortened working memory; the supported outcome is to enable a reviewer to reconstruct evidence after context has been compacted, whereas this scenario requires the ability to avoid answering a pricing question from last month's retrieved cache; a bare compressed claim cannot show which source established it.

Question 306

Before an agent platform release, three separately owned tasks must be covered: one service must meet a requirement that proprietary repository contents remain within an approved execution environment; another must serve many tenants without mixing their tool outputs or prior messages; and a third must retain complete control over a small set of proprietary tool calls and their execution order. Which THREE choices directly address the stated tasks? Choose THREE.

  1. Return a structured tool error to the model
  2. Choose isolated sandbox execution for untrusted code
  3. Use the Claude Agent SDK for an integrated tool-driven agent
  4. Implement a custom Messages API tool loop
  5. Keep user-specific state out of a shared agent process
  6. Deploy agent workers close to controlled data sources

Explanation

Correct Answers

D:

The application receives tool requests, validates inputs, executes authorized functions, and sends tool results back; this directly supports task 3 to retain complete control over a small set of proprietary tool calls and their execution order, not the separate task or tasks to meet a requirement that proprietary repository contents remain within an approved execution environment and to serve many tenants without mixing their tool outputs or prior messages; the Messages API does not automatically execute arbitrary client tools.

E:

Per-request identity and session boundaries prevent cross-tenant context leakage; this directly supports task 2 to serve many tenants without mixing their tool outputs or prior messages, not the separate task or tasks to meet a requirement that proprietary repository contents remain within an approved execution environment and to retain complete control over a small set of proprietary tool calls and their execution order; shared conversation histories can inadvertently expose unrelated customers' data.

F:

A self-hosted harness can enforce infrastructure, network, and retention controls for the surrounding tools; this directly supports task 1 to meet a requirement that proprietary repository contents remain within an approved execution environment, not the separate task or tasks to serve many tenants without mixing their tool outputs or prior messages and to retain complete control over a small set of proprietary tool calls and their execution order; hosted orchestration does not automatically inherit a company's private-network policies.

Incorrect Answers

A:

Tool-result errors surface the failure without impersonating a successful action; this implements the different outcome to recover from a failed search while preserving the agent's chance to choose another path rather than the required tasks to retain complete control over a small set of proprietary tool calls and their execution order and to serve many tenants without mixing their tool outputs or prior messages and to meet a requirement that proprietary repository contents remain within an approved execution environment, and silently dropping a failed tool response leaves the model without a valid observation, so it is not a direct solution to any of the assigned workstreams.

B:

A restricted execution environment limits filesystem, network, and resource privileges; this implements the different outcome to let an agent run generated scripts without exposing production host files rather than the required tasks to retain complete control over a small set of proprietary tool calls and their execution order and to serve many tenants without mixing their tool outputs or prior messages and to meet a requirement that proprietary repository contents remain within an approved execution environment, and running code in the main application process expands the blast radius of a bad tool call, so it is not a direct solution to any of the assigned workstreams.

C:

The SDK provides agent-loop primitives and integrations that remove much of the custom harness work; this implements the different outcome to build a coding agent that needs managed tool execution, sessions, and event handling rather than the required tasks to retain complete control over a small set of proprietary tool calls and their execution order and to serve many tenants without mixing their tool outputs or prior messages and to meet a requirement that proprietary repository contents remain within an approved execution environment, and it still requires explicit permissions, safety policies, and operational boundaries, so it is not a direct solution to any of the assigned workstreams.

Question 307

An inference routing delivery milestone depends on two separate capabilities. One module must call the Messages endpoint from a language without a mature provider SDK; a second module must implement an interactive collaboration channel requiring messages in both directions. Which implementation option is appropriate?

  1. First component: Use direct HTTP REST integration for unsupported runtimes. Second component: Choose SSE for one-way streamed responses.
  2. First component: Use the official SDK for standard requests. Second component: Choose WebSockets for bidirectional realtime exchange.
  3. First component: Use direct HTTP REST integration for unsupported runtimes. Second component: Choose WebSockets for bidirectional realtime exchange.
  4. First component: Choose SSE for one-way streamed responses. Second component: Use the official SDK for standard requests.

Explanation

Correct Answer

C:

The first proposal fits its assigned role because a standards-based HTTP client can send JSON requests with the correct headers and decode responses, while the second proposal fits its assigned role because persistent two-way channels fit interactive event exchange where the client also sends messages during a session; their respective acceptance criteria are to call the Messages endpoint from a language without a mature provider SDK and to implement an interactive collaboration channel requiring messages in both directions.

Incorrect Answers

A:

The first proposal fits its assigned role because a standards-based HTTP client can send JSON requests with the correct headers and decode responses, while the second proposal serves the different purpose to deliver tokens to a browser without maintaining a bidirectional command channel instead of the assigned second role because server-sent events provide ordered server-to-client updates over an HTTP stream; their respective acceptance criteria are to call the Messages endpoint from a language without a mature provider SDK and to implement an interactive collaboration channel requiring messages in both directions.

B:

The first proposal serves the different purpose to simplify authentication, serialization, and response typing in a Python service instead of the assigned first role because maintained client libraries wrap the HTTP API and expose structured response objects, while the second proposal fits its assigned role because persistent two-way channels fit interactive event exchange where the client also sends messages during a session; their respective acceptance criteria are to call the Messages endpoint from a language without a mature provider SDK and to implement an interactive collaboration channel requiring messages in both directions.

D:

The first proposal serves the different purpose to deliver tokens to a browser without maintaining a bidirectional command channel instead of the assigned first role because server-sent events provide ordered server-to-client updates over an HTTP stream, while the second proposal serves the different purpose to simplify authentication, serialization, and response typing in a Python service instead of the assigned second role because maintained client libraries wrap the HTTP API and expose structured response objects; their respective acceptance criteria are to call the Messages endpoint from a language without a mature provider SDK and to implement an interactive collaboration channel requiring messages in both directions.

Question 308

A conversation design application is approaching release, but the team still needs to let an assistant continue a long incident investigation without losing decisions. What is the most appropriate implementation?

  1. Anchor summaries to source identifiers
  2. Compact completed task history
  3. Prune obsolete tool outputs
  4. Prioritize high-signal retrieved passages

Explanation

Correct Answer

B:

A structured summary preserves established facts, pending work, and unresolved assumptions in fewer tokens; this behavior directly supports the need to let an assistant continue a long incident investigation without losing decisions. Raw truncation may remove the reason an earlier decision was made.

Incorrect Answers

A:

When an application needs to enable a reviewer to reconstruct evidence after context has been compacted, stable document IDs and passage references preserve provenance beyond the shortened working memory; the current capability must instead let an assistant continue a long incident investigation without losing decisions, while a bare compressed claim cannot show which source established it.

C:

This approach serves the separate need to stop stale log dumps from crowding out new evidence during an agent session because tool-result pruning removes redundant bulky observations after their findings are retained; it does not establish the outcome to let an assistant continue a long incident investigation without losing decisions, and keeping all raw logs forever does not guarantee better reasoning.

D:

Ranking by relevance and authority places the most useful evidence ahead of incidental matches; despite being useful when a service must fit a large knowledge base into the context of a customer answer, it is not a solution to let an assistant continue a long incident investigation without losing decisions; dumping every document into the prompt can obscure the actual answer.

Question 309

Two features of a production API system are under review. One feature must test whether document classification quality is viable on representative files; the other must exercise an agent's database tool without touching live customer records. Which plan matches the respective needs?

  1. First component: Use a proof of concept before production commitment. Second component: Separate development, staging, and production environments.
  2. First component: Use a proof of concept before production commitment. Second component: Perform a threat review before go-live.
  3. First component: Perform a threat review before go-live. Second component: Instrument production observability.
  4. First component: Instrument production observability. Second component: Separate development, staging, and production environments.

Explanation

Correct Answer

A:

The first proposal fits its assigned role because a small pilot reduces feasibility risk before funding a complete integration, while the second proposal fits its assigned role because environment boundaries keep credentials, data, and writes isolated during testing; their respective acceptance criteria are to test whether document classification quality is viable on representative files and to exercise an agent's database tool without touching live customer records.

Incorrect Answers

B:

The first proposal fits its assigned role because a small pilot reduces feasibility risk before funding a complete integration, while the second proposal serves the different purpose to uncover prompt-injection paths through retrieved webpages and connected tools instead of the assigned second role because reviewing trust boundaries and permissions before deployment reveals attack routes unit tests may miss; their respective acceptance criteria are to test whether document classification quality is viable on representative files and to exercise an agent's database tool without touching live customer records.

C:

The first proposal serves the different purpose to uncover prompt-injection paths through retrieved webpages and connected tools instead of the assigned first role because reviewing trust boundaries and permissions before deployment reveals attack routes unit tests may miss, while the second proposal serves the different purpose to identify rising error rates and latency after launch instead of the assigned second role because logs, metrics, traces, and alert thresholds reveal operational failures at the right layer; their respective acceptance criteria are to test whether document classification quality is viable on representative files and to exercise an agent's database tool without touching live customer records.

D:

The first proposal serves the different purpose to identify rising error rates and latency after launch instead of the assigned first role because logs, metrics, traces, and alert thresholds reveal operational failures at the right layer, while the second proposal fits its assigned role because environment boundaries keep credentials, data, and writes isolated during testing; their respective acceptance criteria are to test whether document classification quality is viable on representative files and to exercise an agent's database tool without touching live customer records.

Question 310

A document processing team is evaluating two disjoint use cases. For one use case, it must let reviewers trace each summary claim to its authoritative document version; for the other, it must prevent a new user from inheriting another user's document excerpts. Which paired choices are justified?

  1. First component: Add explicit provenance to retrieved evidence. Second component: Design an explicit escalation path.
  2. First component: Add explicit provenance to retrieved evidence. Second component: Limit retained context to authorized history.
  3. First component: Design an explicit escalation path. Second component: Add cancellation handling to streaming UI.
  4. First component: Add cancellation handling to streaming UI. Second component: Limit retained context to authorized history.
  5. First component: Keep user tasks in the user message. Second component: Design an explicit escalation path.

Explanation

Correct Answer

B:

The first proposal fits its assigned role because source identifiers, timestamps, and passage boundaries make citations and later audits possible, while the second proposal fits its assigned role because context assembly must apply access checks before adding prior turns or retrieved evidence; their respective acceptance criteria are to let reviewers trace each summary claim to its authoritative document version and to prevent a new user from inheriting another user's document excerpts.

Incorrect Answers

A:

The first proposal fits its assigned role because source identifiers, timestamps, and passage boundaries make citations and later audits possible, while the second proposal serves the different purpose to route uncertain medical-style recommendations to an authorized human instead of issuing final decisions instead of the assigned second role because risk-based thresholds separate assistance from decisions reserved for accountable staff; their respective acceptance criteria are to let reviewers trace each summary claim to its authoritative document version and to prevent a new user from inheriting another user's document excerpts.

C:

The first proposal serves the different purpose to route uncertain medical-style recommendations to an authorized human instead of issuing final decisions instead of the assigned first role because risk-based thresholds separate assistance from decisions reserved for accountable staff, while the second proposal serves the different purpose to stop processing when a user cancels a long report and release resources cleanly instead of the assigned second role because cancellation signals should propagate through streams, queues, and workers where supported; their respective acceptance criteria are to let reviewers trace each summary claim to its authoritative document version and to prevent a new user from inheriting another user's document excerpts.

D:

The first proposal serves the different purpose to stop processing when a user cancels a long report and release resources cleanly instead of the assigned first role because cancellation signals should propagate through streams, queues, and workers where supported, while the second proposal fits its assigned role because context assembly must apply access checks before adding prior turns or retrieved evidence; their respective acceptance criteria are to let reviewers trace each summary claim to its authoritative document version and to prevent a new user from inheriting another user's document excerpts.

E:

The first proposal serves the different purpose to separate an end user's request from the developer's stable operating instructions instead of the assigned first role because explicit message roles distinguish requested work from higher-priority operational framing, while the second proposal serves the different purpose to route uncertain medical-style recommendations to an authorized human instead of issuing final decisions instead of the assigned second role because risk-based thresholds separate assistance from decisions reserved for accountable staff; their respective acceptance criteria are to let reviewers trace each summary claim to its authoritative document version and to prevent a new user from inheriting another user's document excerpts.

Claude Certified Developer - Foundations (CCDV-F) Practice Tests